Training in progress, step 15000, checkpoint

checkpoint-15000/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<bos>": 40478,
+  "<eos>": 40479,
+  "<pad>": 40480
+}

checkpoint-15000/config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "openai-gpt",
+  "afn": "gelu",
+  "architectures": [
+    "OpenAIGPTForSequenceClassification"
+  ],
+  "attn_pdrop": 0.1,
+  "embd_pdrop": 0.1,
+  "id2label": {
+    "0": "entailment",
+    "1": "neutral",
+    "2": "contradiction"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "contradiction": 2,
+    "entailment": 0,
+    "neutral": 1
+  },
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "openai-gpt",
+  "n_ctx": 512,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_layer": 12,
+  "n_positions": 512,
+  "n_special": 0,
+  "pad_token_id": 40480,
+  "predict_special_tokens": true,
+  "problem_type": "single_label_classification",
+  "resid_pdrop": 0.1,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "last",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.0.dev0",
+  "vocab_size": 40481
+}

checkpoint-15000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50cf0c1b5d758ceb5205608f8431621ad7c145001da5891ea2c1662e09ab6cfb
+size 932401285

checkpoint-15000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee91cc89d68e288e5f20afe63584f8cb19c443853b4cb1746222b6df120ff010
+size 466205145

checkpoint-15000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fea505d3bb2cc8e103041c009ac11cec0d896dbfe4c76b96d22f081a5d99d8b8
+size 14575

checkpoint-15000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b4b33c6989ec640db67ac64ffd8c3b5b7340234a656f42a87bafeaadb08649a
+size 627

checkpoint-15000/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<bos>",
+  "eos_token": "<eos>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-15000/tokenizer_config.json

@@ -0,0 +1,7 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 512,
+  "padding_side": "left",
+  "tokenizer_class": "OpenAIGPTTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-15000/trainer_state.json

@@ -0,0 +1,250 @@
+{
+  "best_metric": 1.1149991750717163,
+  "best_model_checkpoint": "/kaggle/output/checkpoint-9000",
+  "epoch": 240.0,
+  "eval_steps": 1000,
+  "global_step": 15000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 2.7777777777777777e-11,
+      "loss": 1.2395,
+      "step": 1
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 2.7638888888888893e-08,
+      "loss": 1.1422,
+      "step": 1000
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.33073852295409184,
+      "eval_loss": 1.1557353734970093,
+      "eval_runtime": 54.1805,
+      "eval_samples_per_second": 92.469,
+      "eval_steps_per_second": 11.572,
+      "step": 1000
+    },
+    {
+      "epoch": 32.0,
+      "learning_rate": 5.541666666666667e-08,
+      "loss": 1.1401,
+      "step": 2000
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.3311377245508982,
+      "eval_loss": 1.1507551670074463,
+      "eval_runtime": 53.98,
+      "eval_samples_per_second": 92.812,
+      "eval_steps_per_second": 11.615,
+      "step": 2000
+    },
+    {
+      "epoch": 48.0,
+      "learning_rate": 8.316666666666666e-08,
+      "loss": 1.1287,
+      "step": 3000
+    },
+    {
+      "epoch": 48.0,
+      "eval_accuracy": 0.3321357285429142,
+      "eval_loss": 1.1441665887832642,
+      "eval_runtime": 53.9873,
+      "eval_samples_per_second": 92.8,
+      "eval_steps_per_second": 11.614,
+      "step": 3000
+    },
+    {
+      "epoch": 64.0,
+      "learning_rate": 1.1094444444444445e-07,
+      "loss": 1.117,
+      "step": 4000
+    },
+    {
+      "epoch": 64.0,
+      "eval_accuracy": 0.3287425149700599,
+      "eval_loss": 1.134506344795227,
+      "eval_runtime": 54.0067,
+      "eval_samples_per_second": 92.766,
+      "eval_steps_per_second": 11.61,
+      "step": 4000
+    },
+    {
+      "epoch": 80.0,
+      "learning_rate": 1.3872222222222222e-07,
+      "loss": 1.1111,
+      "step": 5000
+    },
+    {
+      "epoch": 80.0,
+      "eval_accuracy": 0.32954091816367265,
+      "eval_loss": 1.1283142566680908,
+      "eval_runtime": 54.0921,
+      "eval_samples_per_second": 92.62,
+      "eval_steps_per_second": 11.591,
+      "step": 5000
+    },
+    {
+      "epoch": 96.0,
+      "learning_rate": 1.6650000000000002e-07,
+      "loss": 1.1053,
+      "step": 6000
+    },
+    {
+      "epoch": 96.0,
+      "eval_accuracy": 0.32894211576846305,
+      "eval_loss": 1.1218452453613281,
+      "eval_runtime": 54.0607,
+      "eval_samples_per_second": 92.674,
+      "eval_steps_per_second": 11.598,
+      "step": 6000
+    },
+    {
+      "epoch": 112.0,
+      "learning_rate": 1.9422222222222223e-07,
+      "loss": 1.0952,
+      "step": 7000
+    },
+    {
+      "epoch": 112.0,
+      "eval_accuracy": 0.33073852295409184,
+      "eval_loss": 1.118302583694458,
+      "eval_runtime": 53.9252,
+      "eval_samples_per_second": 92.906,
+      "eval_steps_per_second": 11.627,
+      "step": 7000
+    },
+    {
+      "epoch": 128.0,
+      "learning_rate": 2.2197222222222225e-07,
+      "loss": 1.0909,
+      "step": 8000
+    },
+    {
+      "epoch": 128.0,
+      "eval_accuracy": 0.3281437125748503,
+      "eval_loss": 1.1167163848876953,
+      "eval_runtime": 54.095,
+      "eval_samples_per_second": 92.615,
+      "eval_steps_per_second": 11.591,
+      "step": 8000
+    },
+    {
+      "epoch": 144.0,
+      "learning_rate": 2.4975e-07,
+      "loss": 1.0838,
+      "step": 9000
+    },
+    {
+      "epoch": 144.0,
+      "eval_accuracy": 0.3317365269461078,
+      "eval_loss": 1.1149991750717163,
+      "eval_runtime": 54.0536,
+      "eval_samples_per_second": 92.686,
+      "eval_steps_per_second": 11.6,
+      "step": 9000
+    },
+    {
+      "epoch": 160.0,
+      "learning_rate": 2.775277777777778e-07,
+      "loss": 1.0746,
+      "step": 10000
+    },
+    {
+      "epoch": 160.0,
+      "eval_accuracy": 0.3347305389221557,
+      "eval_loss": 1.1178398132324219,
+      "eval_runtime": 54.1273,
+      "eval_samples_per_second": 92.56,
+      "eval_steps_per_second": 11.584,
+      "step": 10000
+    },
+    {
+      "epoch": 176.0,
+      "learning_rate": 3.0530555555555556e-07,
+      "loss": 1.0722,
+      "step": 11000
+    },
+    {
+      "epoch": 176.0,
+      "eval_accuracy": 0.3373253493013972,
+      "eval_loss": 1.1214321851730347,
+      "eval_runtime": 54.046,
+      "eval_samples_per_second": 92.699,
+      "eval_steps_per_second": 11.601,
+      "step": 11000
+    },
+    {
+      "epoch": 192.0,
+      "learning_rate": 3.3308333333333333e-07,
+      "loss": 1.0617,
+      "step": 12000
+    },
+    {
+      "epoch": 192.0,
+      "eval_accuracy": 0.34271457085828344,
+      "eval_loss": 1.124221682548523,
+      "eval_runtime": 54.0489,
+      "eval_samples_per_second": 92.694,
+      "eval_steps_per_second": 11.601,
+      "step": 12000
+    },
+    {
+      "epoch": 208.0,
+      "learning_rate": 3.608611111111111e-07,
+      "loss": 1.0509,
+      "step": 13000
+    },
+    {
+      "epoch": 208.0,
+      "eval_accuracy": 0.3475049900199601,
+      "eval_loss": 1.1377869844436646,
+      "eval_runtime": 54.0137,
+      "eval_samples_per_second": 92.754,
+      "eval_steps_per_second": 11.608,
+      "step": 13000
+    },
+    {
+      "epoch": 224.0,
+      "learning_rate": 3.8863888888888893e-07,
+      "loss": 1.0381,
+      "step": 14000
+    },
+    {
+      "epoch": 224.0,
+      "eval_accuracy": 0.34890219560878244,
+      "eval_loss": 1.155613899230957,
+      "eval_runtime": 53.9794,
+      "eval_samples_per_second": 92.813,
+      "eval_steps_per_second": 11.616,
+      "step": 14000
+    },
+    {
+      "epoch": 240.0,
+      "learning_rate": 4.163888888888889e-07,
+      "loss": 1.0255,
+      "step": 15000
+    },
+    {
+      "epoch": 240.0,
+      "eval_accuracy": 0.3562874251497006,
+      "eval_loss": 1.1687414646148682,
+      "eval_runtime": 53.8948,
+      "eval_samples_per_second": 92.959,
+      "eval_steps_per_second": 11.634,
+      "step": 15000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10000000,
+  "num_train_epochs": 161291,
+  "save_steps": 1000,
+  "total_flos": 3.135532695552e+16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-15000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb2cd1b7e648a386ccb79e47a451842e11001981c4363e993b678e0818db5cab
+size 4091